Two-stage object detection that combines a passive infrared sensor with an ultrasonic ranging sensor is widely adopted in low-cost room security devices, yet most published prototypes report only functional success without characterising the metrological behaviour of each sensing stage. This study characterises and calibrates the sensing subsystem of a room monitoring prototype built around a Wemos D1 R32 microcontroller, an HC-SR501 passive infrared sensor, an HY-SRF05 ultrasonic sensor, an SG90 servo motor, and an active buzzer. An experimental quantitative approach was applied: motion detection was tested at six standoff distances up to 240 cm, ultrasonic ranging was compared against a reference ruler at five distances from 20 cm to 100 cm, and the servo sweep period was measured over five replications. Percentage error, ordinary least squares regression, and descriptive statistics were used in the analysis. The passive infrared stage detected motion consistently up to 200 cm and failed at 240 cm, while the buzzer responded within 1.30 s to 2.64 s. The ultrasonic stage produced a mean relative error of 1.16%, dominated by a fixed positive offset of 0.96 cm; a two-parameter linear calibration with a coefficient of determination of 0.9999 lowered the mean relative error to 0.47%. The servo completed a 360 degree round sweep in 5.446 s, matching the blocking delay budget of the firmware. The findings demonstrate that detection latency is governed by firmware architecture rather than sensor physics, and that a simple linear calibration substantially improves ranging accuracy.